Pressure Development in Dust Explosions - An experimental study in the 20-liter Siwek sphere and a 0.5 m³ non-standardized vessel
MetadataShow full item record
This thesis describes an experimental investigation of pressure development in dust explosions, using two different volumes. The 20-liter Siwek sphere is a standard explosion test apparatus used for determining maximum pressures and maximum rate of pressure rise of an explosion. The other vessel, with a volume of 0.5 m³ is built from scratch for the same purpose. The aim is to see whether the same explosion properties are obtained in the two vessels. The scaling law (‘cube-root law’) at which one determines the severity of an explosion and evaluate mitigation safety measures after, is said to be independent of vessel size. One could therefore expect the same explosion severities in both vessels. Conversely, it is known that several parameters such as chemical composition, particle size distribution and turbulence level will influence a dusts’ explosion violence. Organic dust holding two different moisture contents as well as aluminum dust were tested in both vessels in accordance to the European Standard, and evaluated. In addition, results acquired by GexCon AS from a 25 m³ vessel were considered and compared to the explosion properties obtained in this work. It was relevant to investigate validity of the ‘cube-root law’ with regard to combustion of dusts where thermal radiation is thought to be important, which is typical for aluminum. Thermal radiation can heat up particles faster and farther away from the reaction zone than thermal convection. This may lead to an increase of flame propagation velocity with radius, and thus the explosion violence will indeed be affected by vessel size. The results showed some deviations in explosion violence when increasing the vessel volume, as did different particle size distributions. However, large discrepancies were observed when altering moisture content and turbulence levels at the time of ignition. Thus, the K_St-value seems to be an approximate measure of explosion violence, and should not be blindly trusted when designing mitigation safety measures in the industry.